Jar Jar Binks explains shell commands to you via GPT-3:

Aug 26, 2022 · 2:32 AM UTC

“Let’s think step-by-step” is also Jar-Jar-generalizable:
Style of output improves when answer prefix demonstrates proper usage of "meesa", substitution of dental fricatives:
Replying to @goodside
How do I play with this tool?
These are screenshots from OpenAI Playground.
Replying to @goodside
I find that the probability spaces around statements of fact rather than commands or requests are much more effective. Interacting with it as if it’s an entity narrows the spaces to interactions instead of intuitive thoughts. So, “The following is” rather than “Write an XYZ…”
I go back and forth between the two. I find for InstructGPT, the declarative “This is …:” style also causes it to ramble on after the intended output instead of stopping on its own, so you need to define and use stop sequences. Makes it harder to try spontaneous ideas.
Replying to @goodside
Off topic, but is there a prompt that only answer with abbreviations (to save tokens and better utilize the 4k limit)
To reduce token length of generations, consider fine-tuning. You can include whatever abbreviated output format you want in your fine-tuned examples and it should learn it. But you need a decent number of examples for it to work.